Latest
  1. Johns Hopkins Bayview tells 3,182 patients an employee put their health data into a public AI tool
  2. China’s AI compute access expected to double in 2027, with US-designed chips about 65 per cent
  3. UK commits £100m to national AI Biosecurity Lab
  4. Moonshot AI releases Kimi K3.1 Pro with 1-million-token context window
  5. Residents in Newport and Cardiff oppose hyperscale datacentres near homes
  6. Study: Grok 4.7 picked Republicans in 92% of forced midterm answers, GPT-6.1 Sol picked Democrats in 97%
  7. Attackers exploiting the PHP flaws hand the break-in to AI agents, Unit 42 says
  8. Elvis Presley’s rightsholders sue Anthropic over training on 274 songs

Latest / Frontier models and safety

ReportFrontierScienceUSConfirmed

GPT-6 Astra jumps to 62.7% on ARC-AGI-3, five months after models scored 0.5%

See how this story connects

Source: ARC PrizeAlso reported by ARC Prize (ARC-AGI-3 launch)

ARC Prize launched ARC-AGI-3 on 25 March 2026 as hundreds of instruction-free interactive games where humans scored 100% and frontier models 0.51%. On 3 September it reported OpenAI’s GPT-6 Astra at 62.7% (about $26,000) on its standard harness and 99.9% using OpenAI’s own context-management adapter, using fewer actions than the human baseline on 96% of levels. ARC cautioned that the benchmark has closed-ended games and does not represent the open-endedness of the real world.

Why it matters

The same model scored 62.7% on ARC’s standard harness and 99.9% with OpenAI’s own adapter, so anyone comparing ARC-AGI-3 results needs to know which setup produced them.

When
3 Sep 2026
ID
WTA-N-2026-0082
Who
ARC Prize Foundation, OpenAI

Summary by wheresthe.ai, from ARC Prize and the reporting credited above. Checked against the source on 7 Oct 2026.

Read the original at ARC Prize

Source copy: download the page as saved on 18 Oct 2026 (PDF). Also held by the Internet Archive.

Earlier in this Storyline: What led hereReview of 445 LLM benchmarks finds widespread construct-validity weaknesses

Cite this story

WTA-N-2026-0082: “GPT-6 Astra jumps to 62.7% on ARC-AGI-3, five months after models scored 0.5%.” wheresthe.ai, 3 Sep 2026. https://wheresthe.ai/d/arc-agi-3-gpt-6-astra-2026-09/ (source: ARC Prize).

Spotted an error? Email [email protected].